Re: Accurately measuring current draw

[michalwd1979]

Hello Faris,
I have used clamp meters in my normal work. The models I used were really accurate. In my opinion these points are important:
1) You don't need to measure DC current so don't buy one that let's you measure it. They are much more expensive. Get the best AC one that you can afford.
2) PFC corrected power supplies should not produce much distortion in current but true rms meter is always a good idea. RMS current trips breakers and heats up wires
3) Whatever power factor yours servers will have the current is the limit of the installation. If Yours main point is to be sure that you are within the limits you don't need a real power meter. Simple current measurement should be perfectly OK
4) Servers and UPSes typically have large inrush currents. It can trip some fuses and it is very hard to measure it accurately. Once (not in server  room of course) I had to use scope with memory to prove that inrush current is the problem. If You have upses with many individually controlled outputs try to power up them in sequence after power failure.
5) I never used Kill-A-Watt, so I can not make any comments about it.

I am not sure that the above is 100% correct, I'm not a sysadmin.
Hope this helps
Michael


Dnia 5 marca 2008 14:30 "Faris Raouf" <asterisk at raouf.net> napisał(a):

> Following on from the discussions on the Kill-A-Watt in another thread, has
> anyone used a proper clamp meter for measuring current?
> 
> I'm thinking about buying a cheap-ish one ($80/£40), along with a plug-in
> device similar to the Kill-A-Watt in order to compare these two options
> (which I consider "slightly-pro" and "consumer" - the clamp meter is only
> slightly-pro because it isn't an expensive one - no calibration certificate
> etc.
> 
> But I'd appreciate any feedback on how accurate the Kill-A-Watt type
> products might be before I bother doing so.
> 
> What I mean by accurate in this sense is given, say, an 8A power feed in a
> data centre, I want to be 100% sure that if my test figures say each server
> takes less than 1A under extreme load and I have no more than 8 of them I'm
> not going to trip the breaker.
> 
> If there's more to this than meets the eye (e.g. all that RMS stuff and god
> knows what else that I've forgotten about since I did my Physics) please can
> someone pipe up and let me know?
> 
> 
> Faris.
> 
> 
> 
> 
> 
> _______________________________________________
> Linux-PowerEdge mailing list
> Linux-PowerEdge at dell.com
> http://lists.us.dell.com/mailman/listinfo/linux-poweredge
> Please read the FAQ at http://lists.us.dell.com/faq
>